Welding Tips and Tricks

I have a some large aluminum diesel fuel tanks to repair. I don't have a problem with the welding but the cracks are all in the same place and i'm wondering what the problem is. On the bottom there is a small formed plate with the outlet and return line couplngs welded in. This plate is inset and welded to the tank. All the tanks have cracks in this plate or in the weld connecting them to the tank. I'm thinking maybe the lines inside may have broken loose from some type of brace and vibration is causing the problem. Anyone seen this situation before?

Welcome to the forum. I have welded quite a few alum fuel tanks and have never personally seen what you are describing. Off hand it sounds like there is stress causing the cracks? Can you get up a picture to help us out?
-Jonathan

Hope i did the picture correctly.

OK. picture worked. This part is not prepared for welding yet. The cracks are all either in the weld or in the formed insert itself. It varies.

If what you suspect is true this bad boy might just be the ticket
to confirm your suspicions.

http://www.harborfreight.com/digital-in ... 67979.html

Nathan,
Are the tanks hard-plumbed to the fuel line, or is there a rubber section of hose to isolate shock and vibration?

There are rubber hoses that connect to the fittings. Thanks, that inspection camera will keep me from cutting the end off the tank to look.
I appreciate the help guys, I have about 16 of these to repair.

To have 16 with the same problem in the same area, it sounds like a design flaw.

Might try contacting the manufacturer to see if there's an engineered fix for it already. Hell, there might be a "recall", where the maker will pay you for the fix, or at least a kit.

I suspect a materials mis-match, like 3003 for the body of the tank, and 6061 for the insert, or too hard a filler metal for the weld, that's work-hardening or age-hardening.

Just possibilities to consider...

Steve S

I've had a couple of similar tank jobs - but these had had a hard life & were knocked around bad anyway so it was never determined what the cause was - but the crack was around a similar deeply pressed part - the owner did'nt care how it was repaired & I suspected it may have been a combination of the weld ( visually looked like synergic pulse mig - around here almost always done with 5356 due to feeding concerns ) & high stresses produced by the deep pressed shape - tank alloy was not known - my repair method was to roll a piece of much thicker aluminum to the tank radius & as it was already shaped like an inspection port on a pressure vessel I inserted the new piece inside the tank then drilled a series of holes around it - removed the piece - tapped them & installed it with screws & a little sealer - not had any problems since & if the tank gets much crap in it can be removed for easy cleaning - owner was happy.

It seems to me that this panel to which the insert is welded might be flexing due to change in temperature or fuel level causing the weld or insert to work harden.
If the manufacturer does not have a fix for you, maybe some stiffening angles welded to the tank near both sides of the insert would take the flex out of the panel and protect the insert.

I recently repaired about a dozen, 300 gallon portable diesel tanks that fit inside the back of pickup trucks for disaster relief. Most of them cracked in the same place. I later figured out that they were being shoved out of the trucks and let land on their ends after being returned.

Another series of tanks, had been fabricated with somewhat centrally located fittings, to where a lot of distortion was welded into the top panels, causing a rather severe oil canning effect when either filled quickly with fuel, or change in temperature as mentioned above. The rosette welds that had been used to subdue the bulkhead/slosh plates were also cracking. The top panel with the fittings was exchanged with panels that were cross broken and the fittings moved closer to the corners.

First thing I did was repair all of the leaks. Then pressure tested the tanks to about 3-5 psi to check movement. It was substantial.

Poor weld penetration in poorly fitted open corner joints. I tack mine and tweak them accordingly to make sure the reveal is even across the joint. These had been allowed to overlap in places and the welders had just tried to burn them in.


During pressure testing, the deformed sections move as much as an inch, and rather abruptly. You could visibly see the effect it would have on the welds over time.


So basically, we subdued the movement and corrected the welds. I had to grind the welds back to where I could see the corners, which a clean line of unwelded interior corners of the sheet edges were apparent. I try to use just the right heat and speed to lay a well convex'd bead, with the edges of the weld just a RCH (on the verge of being overlap) above flush. No undercut, or heat fatigue allowed.

I'd rather use self fabricated NPT fittings, perhaps pre-installed on a doubler made of the tank material than prefab.